Lance Herron

450 posts

Lance Herron banner
Lance Herron

Lance Herron

@theLance

Relapsed SWE. Claude whisperer

TX Beigetreten Eylül 2008
528 Folgt81 Follower
Lance Herron
Lance Herron@theLance·
@nisten Need them to hurry up and drop M5 Mac Studio!
English
0
0
2
24
nisten🇨🇦e/acc
it's really starting to look like local AI at home is crossing the practicality/price ratio ppl actually running agents on consumer hardware now...
Naz Bent@Bent302

Benchmarked @DJLougen ’s Ornstein-27B-v2 Q6_K on my RTX 3090 using hermes-bench, my new open-source benchmarking UI for local LLMs and Hermes agents. Ornstein is a Qwen 3.5 27B fine-tune trained on reasoning traces filtered through a Drift Diffusion Model pipeline. Quality over quantity. The DDM separates “fake” reasoning (hedging, restating, circling) from the real thing with >99% sensitivity. Running llama.cpp + TurboQuant turbo3_tcq KV compression. LLM-as-judge scoring via Carnice-9b. Real tool calls, real execution, no synthetic evals. 12 tasks across two suites. Results thread below. 🧵 Model: huggingface.co/DJLougen/Ornst…

English
2
1
4
283
Lance Herron retweetet
Uncle Bob Martin
Uncle Bob Martin@unclebobmartin·
AIs aren’t good rule followers. The older the rule in the context window, the less priority it is given. So the best way to enforce the rules is with external tools that communicate failure to the AI. Acceptance testers, Linters, dependency checkers, C.R.A.P. analysis. Mutation testing. Etc. Productivity gains come from disengagement from the code. Let the AI worry about the code. You worry about everything else, including the code quality metrics.
English
12
19
188
9.2K
Lance Herron
Lance Herron@theLance·
@lessin …or you can sign up 3x the users using the same amount of compute. Every user gets the tiny-bit-dumber version of the model, nobody cancels because it’s still better than nothing, and you have almost returned to zero marginal cost per user.
English
0
0
0
15
sam lessin 🏴‍☠️
”Cognitive Shaving” / Debasement is the new Ad-Load Dial -- If Your Quarter is Light?… Just be a 0.1% dumber, or spend 0.1% more tokens — whatever you need to hit your goal.
sam lessin 🏴‍☠️ tweet media
English
6
2
12
1.8K
Lance Herron
Lance Herron@theLance·
@emollick This benchmark looks like “METR Time Horizon for Cyber”. Not saying concern is unwarranted, but is it really unexpected? Where is the benchmark for how capable Mythos is at fully testing/securing a network?
English
0
0
0
207
Lance Herron
Lance Herron@theLance·
@iruletheworldmo Codex CLI is way behind Claude Code CLI. They just don’t have the shipping velocity of CC. Rust may have been a mistake. Spud may be a great model but it will be hobbled if they don’t fix the harness.
English
0
0
0
288
🍓🍓🍓
🍓🍓🍓@iruletheworldmo·
openai has obviously seen the money claude code is pulling in they’ve pivoted hard codex ain't no side project it’s becoming the main thing the $100 plan was the first tell next week is probably where that really starts to show up, they have big plans ^^ my guess: more codex plans more codex surface area and maybe the first public taste of the bigger stuff too spud new image model whatever else they’ve been hiding behind all this vagueposting either way i think next week is when codex stops looking like a tool and starts looking like the center of the product from what im hearing the spud will be incredibly strong. they've made some novel breakthroughs. i'm very excited.
English
45
12
433
29K
Lance Herron
Lance Herron@theLance·
@MatthewBerman @beffjezos The best setup is using Opus to drive Codex. You get the prototypical “cracked but socially inept” engineer without having to talk to him.
English
0
0
4
239
Matthew Berman
Matthew Berman@MatthewBerman·
@beffjezos Everyone seems to say this but I seem to always go back to Opus 4.6
English
29
0
106
7.9K
Beff (e/acc)
Beff (e/acc)@beffjezos·
When you've been too locked in on Claude and finally try out GPT 5.4 high for a coding task only to realize what you've been missing out on for weeks...
GIF
English
157
61
1.8K
130.1K
Lance Herron
Lance Herron@theLance·
@emollick Thought about asking ChatGPT for examples of these, but decided against it. I don't think I want to see the things that can't be unseen.
English
0
0
0
130
Ethan Mollick
Ethan Mollick@emollick·
Chiasmus (reversing grammatical structures in two sentences for drama). Asyndetic tricolon (three items listed without a conjunction). Parataxis (short and somewhat disconnected dramatic sentences). Same stuff in every post and essay. Once you see it, it is everywhere.
English
12
6
92
12.4K
Ethan Mollick
Ethan Mollick@emollick·
A lot of our education on writing well focuses on logic, clarity, and argument. AI will force us to think more about style. The boredom that comes from everything on the internet reading Claude-y now, no matter how good the substance is, should make us appreciate variety more.
English
41
43
434
30.7K
Lance Herron
Lance Herron@theLance·
Reminder: Weekends are for cleaning up and culling all the slop-code you generated during the week.
English
0
0
0
4
Lance Herron
Lance Herron@theLance·
@RayFernando1337 For me the Claude app is basically just a remote renderer for Claude Code at this point. Just spin up a few remote-control instances in the morning and avoid the mobile limitations.
English
0
0
1
313
Ray Fernando
Ray Fernando@RayFernando1337·
Opus 4.6 Extended chat on iOS is capped at 10k tokens for thinking which makes me burn more tokens for the same task. I’ve noticed the model used to take a lot longer to process my requests and it would do multiple tool calls to get work done the first time. Now I have to keep prompting the model multiple times and I don’t get the same outcome. It feels like the model is dumb because it makes too many tradeoffs and ends up wasting my time.
Ray Fernando tweet media
English
16
6
190
28.5K
Lance Herron
Lance Herron@theLance·
Now I have the full picture.
English
0
0
0
2
Lance Herron
Lance Herron@theLance·
@noahzweben Ok..not so awesome. Now any Bash calls that use sleep error out. Opus doesn't seem smart enough to use Monitor tool (or it's not available yet) so it backgrounds all the polling and churns through thousands MORE tokens. Does not seem well thought out.
English
0
0
1
13
Lance Herron
Lance Herron@theLance·
@noahzweben This is awesome. How do we get visibility into what triggers a turn? Or how do we steer it on what we want to trigger a turn? Some tool call hook shenanigans would be cool here!
English
1
0
2
1.7K
Noah Zweben
Noah Zweben@noahzweben·
Thrilled to announce the Monitor tool which lets Claude create background scripts that wake the agent up when needed. Big token saver and great way to move away from polling in the agent loop Claude can now: * Follow logs for errors * Poll PRs via script * and more!
English
232
468
6.2K
1.2M
From Laniakea
From Laniakea@FromLaniakea·
@ThePrimeagen *only to select customers **while gpu supply lasts ***terms apply (but we won't tell you which)
English
1
0
8
791
ThePrimeagen
ThePrimeagen@ThePrimeagen·
mythos is coming
ThePrimeagen tweet media
English
90
30
1.6K
43.4K
Lance Herron
Lance Herron@theLance·
@Noahpinion I personally prefer AI takes over admin takes. On the forecasting stuff, I wonder if the respondents were considering scale and tenacity. Millions of average-skill researchers running 24/7 could lead to outcomes similar to their 14% scenario.
English
1
0
1
528
Lance Herron
Lance Herron@theLance·
@steipete I’ve been working on personal projects this way for a few months. Adding one more layer of indirection above the orchestrator (a task/process manager) made it even better.
English
0
1
0
208
Peter Steinberger 🦞
Peter Steinberger 🦞@steipete·
Working on a new QA process where we use openclaw to QA openclaw with a new synthetic message channel. Orchestrator agent understands project, defines tasks (e.g. ask agent to create cron) then verifies that agent did what that. If failure -> spin up subagent to analyze+fix.
English
76
30
901
58.7K
Lance Herron
Lance Herron@theLance·
@thdxr @dexhorthy I’m sure we’ll get some nice “Oh sorry we’re working on clarifying the messaging” tweets shortly.
English
0
0
0
180
dax
dax@thdxr·
@dexhorthy i was a bit annoyed that people were buying the blame shift of "third party harnesses don't call our apis right" if that was the real reason there's like 3 relevant ones and it's easy to work with them to get what you need
English
9
0
140
8.5K
dex
dex@dexhorthy·
Concerning re: anthropic. The previous narrative just went out the window Reports of openclaw usage with the plain sdk (as was supposedly permitted) now being blocked based on system prompt, even if using the claude agent sdk I was previously a little to the anthropic side of the spectrum on this because of EXACTLY one argument - “third party harnesses don’t use caching properly, can’t be controlled with feature flags, etc” If they are blocking use of the claude agent sdk wholesale in openclaw, then this completely invalidates that argument and I desire an answer as to what is allowed and why. I am disappointed that the communications thus far have failed to articulate the reasons here, and does make it harder to trust whatever they say next. However I will maintain cautious optimism that there is a good explanation for all this beyond the cheap “rug pull” “evil” “kill all the startups” jeers
dex@dexhorthy

like I’ve said a few times, well within TOS to do this, they built the model, if they wanna give you inference at pennies on the dollar on the condition that you use their harness, great, they have the right to do this. On this topic in particular, I don’t understand the “evil” or “rugpull”, jeers. There was never any promise to give people cheap inference. Before the claude code max plan we were all paying per token to use this stuff. And we’re more or less happy to do it (sure the VC funding helps). Every enterprise I know pays per token because when you use subsidized inference, YOU are the product. “Have some cheap code, in exchange for helping to train the next gen of models” You can hate on that particular behavior if you want but nobody is making you take part in that particular market dynamic. Do I wanna see a world where model companies take some of their massive financial gains and use that to pull everybody up? Of course. I hope it happens some day. An allegory perhaps: If public e-bike company gave you a subscription on rides and you proceeded to around ripping out batteries and sticking them in your own bike and ride around town, you’d get banned for that too. Especially if your bike was poorly wired and overloaded the batteries/cause them to flame up etc. Banning that behavior would deliver far better results for the people who were using the system as designed

English
52
26
442
144.3K
Lisan al Gaib
Lisan al Gaib@scaling01·
Claude Mythos is really slow and stupid today... they nerfed it again, didn't they?
English
27
1
363
58.8K